Radiolabelled compounds may be used as radiopharmaceuticals in a number of applications such as in radiotherapy or diagnostic imaging. In order for a radiolabelled compound to be employed as a radiopharmaceutical there are a number of desirable properties that the compound should ideally possess such as acceptable stability and, where possible, a degree of selectivity or targeting ability.
Initial work in the areas of radiopharmaceuticals focussed on simple metal ligands which were generally readily accessible and hence easy to produce. A difficulty with many of these radiolabelled compounds is that the complex formed between the ligand and the metal ion was not sufficiently strong and so dissociation of the metal ion from the ligand occurred in the physiological environment. This was undesirable as with the use of ligands of this type there was no ability to deliver the radiopharmaceutical to the desired target area in the body as metal exchange with metal ions in the physiological environment meant that when the radiopharmaceutical compound arrived at the desired site of action the level of radiolabelled metal ion coordinated to the compound had become significantly reduced. In addition where this type of exchange is observed the side effects experienced by the subject of the radiotherapy or radio-imaging are increased as radioactive material is delivered to otherwise healthy tissue in the body rather than predominantly to its place of action.
In order to overcome the problem of metal dissociation in the physiological environment a number of more complicated ligands have been developed and studied over time. Thus, for example a wide range of tetra-azamacrocycles based on the cyclam and cyclen framework have been investigated. Examples of ligands of this type include DOTA and TETA.

Unfortunately, even with these ligands there is still dissociation of the metal with certain derivatives. For example, some derivatives suffer from dissociation of Cu from the chelate as a consequence of transchelation to biological ligands such as copper transport proteins either as Cu2+ or following in vivo reduction to Cu+.
In order to increase the stability of radiolabelled compounds therefore hexaminemacrobicyclic cage amine ligands, known by their trivial name sarcophagines have been developed. These cage ligands form remarkably stable complexes with metals such as Cu2+ and have fast complexation kinetics even at low concentrations of metal at ambient temperatures. These features therefore make ligands of this type particularly well suited in radiopharmaceutical applications, especially those applications involving copper.
Once the problem of stability of the complex between the ligand and the metal had been overcome attention turned to developing ways in which the ligand could be functionalised to incorporate targeting molecules within the ligand without compromising the stability of the metal ligand complex or the ultimate biological activity of the targeting molecule. A number of different targeting molecules are known in the art and the issue became how best to attach these to the ligand molecules.
In general the targeting molecule (or molecular recognition moiety as it is sometimes known) is attached to the ligand to provide a final compound containing both a ligand and a molecular recognition moiety. Whilst these compounds may contain a single molecular recognition moiety they may also be multimeric constructs where the ligand is attached to two (or more) molecular recognition moieties. This is typically desirable as a multimeric construct can possess higher affinity for a target receptor than its monomeric equivalent. This is in part due to an increase in the local concentration of the targeting group, allowing it to compete more effectively with endogenous ligands. In addition in circumstances where there is sufficient length between two or more targeting groups within a multimeric construct, then cooperative binding is possible, and two or more targeting groups will bind to two or more receptor sites at the same time. Indeed it has been observed that in vivo, a multimeric construct often demonstrates higher target tissue accumulation than its monomeric equivalent. Without wishing to be bound by theory it is thought that this is due to the higher affinity of the multimeric construct for the target receptor than that of the monomeric construct. Furthermore, the multimeric construct has a higher molecular weight than the monomeric construct and therefore prolonged bioavailability (as it is more resistant to degradation in the physiological environment). This can result in increased accumulation and retention in target tissue.
Initial work in the caged ligand area looked at direct coupling reactions of the primary amines of the cage amine ‘diaminosarcophagine’, 1,8-diamino-3,6,10,13,16,19-hexaaza bicyclo[6.6.6] icosane ((NH2)2sar), with peptides using standard coupling procedures. Unfortunately for a variety of reasons this has proven to be relatively inefficient and work in this area ceased. Workers then focussed on the incorporation of an aromatic amine to produce SarAr. The pendent aromatic amine can be used in conjugation reactions with the carboxylate residues of peptides or antibodies and it has been shown that SarAr could be conjugated to anti-GD2 monoclonal antibody (14.G2a) and its chimeric derivative (ch14.8) and the conjugate has been radiolabelled with 64Cu.

A difficulty with this approach is that in reaction of the aromatic amine in the conjugation step there are 8 other nitrogen atoms in the SarAr molecule that are available for competing reactions leading to the potential for the creation of a large number of impurities that is undesirable from a pharmaceutical sense. Whilst these could potentially be overcome by the use of substantial protective group chemistry this is clearly undesirable from a synthetic standpoint and scale up on a commercial scale.
An alternative approach has been to elaborate the ligand to incorporate carboxylate functional groups and incorporate peptides or antibodies via their N-terminal amine residues and this approach is of particular importance when the C-terminus is crucial to biological activity. Studies have shown that (NH2)2sar, can be functionalised with up to four carboxymethyl substituents via alkylation reactions with chloroacetic acid and the introduced carboxymethyl arms can be used as a point of further functionalisation and EDC-coupling reactions can then be used to introduce amino acids.
Unfortunately a potential disadvantage of these systems is that intramolecular cyclisation reactions can still occur in which the carboxymethyl arm reacts with a secondary amine of the cage framework to form lactam rings resulting in quadridentate rather than sexidentate ligands. Accordingly whilst this approach can be followed the potential for unwanted side reactions is clearly undesirable from a commercial perspective.
Further studies directed towards the functionalisation of ((NH2)2sar) were based around its reaction with activated di-carbonyl compounds such as acid anhydrides leading to the formation of an amide bond to the amine nitrogen and a free carboxylic acid moiety which was available for further elaboration to the desired binding onto a molecular recognition moiety. A difficulty with this approach is that these reactions typically involve the use of coupling reagents which, given the intended end use of the compounds, may be undesirable especially if any residue is hard to remove. Accordingly there remains a need to develop new and improved methods of functionalising ligands of this type.